Thesis (Ed. D.)--Illinois State University, 1982. / Title from title page screen, viewed April 4, 2005. Dissertation Committee: Robert K. Rittenhouse (chair), Lanny E. Morreau, Ronald Halinski, Carol Mardell Czudnowski, Janet Hartman. Includes bibliographical references (leaves 107-110) and abstract. Also available in print.

No description available.

This study examined the effect of intervention involving a humanoid robot on challenging or tantrum behaviors of four children with low-functioning autism spectrum disorders (ASD). The current work was part of a larger study involving the effect of the robot on functional communication skills on a variety of different interactions with different communication partners. All participants took part in a single-subject, multiple-baseline design with various session types including baseline, traditional play-based treatment, treatment including the robot, and follow-up sessions. For the purpose of this study, only the sessions including treatment with the robot where the robot interaction occurred at the beginning or the end of a 50-minute session were analyzed. Six different categories of tantrum behaviors were analyzed during the sessions including: crying/screaming; self-distracting behaviors; biting, hair pulling, squeezing or pinching; throwing/shoving; and hitting/kicking. Results indicate that for 3 out of the 4 children, tantrum behaviors decreased when the robot interaction came at the beginning of the session. The fourth child showed minimal change in tantrum behaviors. To improve understanding of the influence of a robot on children with ASD, future research should be conducted to determine what aspects of the robot interaction would be most effective on reducing these behaviors.

Multilevel modeling represents a potentially viable method for meta-analyzing single-subject research, but questions remain concerning its methodological properties with regard to characteristics of single-subject data. For this dissertation, Monte Carlo methods were used to investigate the properties of a 3 level model (i.e., with a quadratic equation at level 1), and three different level 1 error specifications (i.e., different variance components and covariances of 0, lag-1 autoregressive covariance structures, and separate error terms for each phase, with different variance components and covariances of 0). Data for simulated subjects were generated to have characteristics typical of published single-subject data (e.g., typical variances and magnitudes of effect). Samples were simulated for conditions which varied in number of data points per phase, number of subjects per study, number of studies meta-analyzed, level of autocorrelation in residuals, and continuity of variance across phases. Outcome variables examined included rates of convergence of analyses, power for statistical tests of fixed effects, and relative parameter bias of estimates of fixed effects, random effects’ variance components, and autocorrelation estimates. Convergence rates were found to be 100% for all level 1 error specifications and data conditions. Power for statistical tests of fixed effects was observed to be adequate when 10 or more data points were generated per phase and 60 or more total subjects were included in meta-analyses. The relative biases of estimates of fixed effects were found to have limited associations with numbers of data points per phase, levels of autocorrelation, and the continuity/discontinuity of variance across phases. Random effects’ variance components were observed to be frequently biased. Associations between relative bias and data conditions were found to vary by random effect. Finally, autocorrelation estimates were found to be biased in all conditions for which autocorrelation was generated. Results are discussed with regard to study strengths and limitations, and their implications for the meta-analysis of single subject data and primary single subject research. / text

While enrolling in a college for a degree is still considered to be more of the expectation than the rule in households across the United States it is facing public criticism regarding its overall cost versus long-term benefits. This has prompted researchers to reevaluate practices in higher education to determine ways in which post-secondary institutions can improve their overall outcomes. In part, research indicates that this should include revisions to assessment, evaluation, and research (AER) practices. Currently, higher education faces many challenges associated with rigorous AER practices, which include an institutional focus on accountability measures, organizational challenges, and difficulties in operationally defining and measuring constantly changing definition for students, faculty, and higher education. These challenges have forced higher education to move away from rigorous quantitative designs that show causality. Additionally, higher education’s focus on large N research struggles to capture the unique identities and experiences of today’s college campus, which also poses challenges for the use of AER by practitioners across academic and student affairs. At present, new approaches to AER must be considered. The purpose of the current study was to propose and investigate one potential way to supplement large N AER with the use of single-subject research design (SSRD). Using visual analysis and the calculation of effect sizes with nonoverlap of all pairs (NAP), archival data from a unique behavioral program in a post-secondary transition program located at a land-grant institution in the southeastern United States was evaluated to determine if its implementation reduced student referrals at the program (N =25) and cohort level between Fall 2019 to Spring 2020. Results indicated that the behavioral program was most effective at the program level but had variable results at the cohort level. These results indicate that SSRD can be an effective approach to AER practices in higher education and would adequately supplement and potentially inform further large N research with its ability address to assess smaller populations.

This study contributes to the existing literature on treatment integrity (TI) by presenting TI findings across interventions aimed at the development of social emotional skills in learners with emotional and behavioural disorders. Social and emotional skills were selected as the target for our investigation given the significance of these skills in relation to the academic and behavioural success of learners and the challenges most often faced in these skill areas by students with emotional and behaviour disorders (E/BD). The study analysed single-subject experimental studies from 2000 to 2009 in two leading journals in the field of emotional and behaviour disorders: Behavior Disorders and The Journal of Emotional Behavioral Disorders. The degree to which studies operationally defined independent variables and evaluated and reported measures of treatment integrity and associated risk factors is reported. Thirty-three studies met the inclusion criteria for the present study and TI was evaluated across six variables (1) year published, (2) dependent variable(s), (3) independent variables(s), (4) participant characteristics, (5) treatment agent and (6) assessment of TI. Results indicated that approximately 49% of the studies monitored and reported TI, meaning that they provided a description of the TI procedure and resultant data. Findings from the study point to the need for attention to TI both in the description of methods used and in the reporting of TI data.

Athletes who merit the title ‘elite’ are rare and differ both quantitatively and qualitatively from athletes of lower qualifications. Serving and studying elite athletes may demand non-traditional approaches. Research involving elite athletes suffers because of the typical nomothetic requirements for large sample sizes and other statistical assumptions that do not apply to this population. Ideographic research uses single-athlete study designs, trend analyses, and statistical process control. Single-athlete designs seek to measure differences in repeated measurements under prescribed conditions, and trend analyses may permit systematic monitoring and prediction of future outcomes. Statistical process control uses control charting and other methods from management systems to assess and modify training processes in near real-time. These methods bring assessment and process control into the real world of elite athletics.

This study presents a set of data analysis approaches for single subject designs (SSDs). The primary purpose is to establish a series of statistical models to supplement visual analysis in single subject research using Bayesian estimation. Linear modeling approach has been used to study level and trend changes. I propose an alternate approach that treats the phase change-point between the baseline and intervention conditions as an unknown parameter. Similar to some existing approaches, the models take into account changes in slopes and intercepts in the presence of serial dependency. The Bayesian procedure used to estimate the parameters and analyze the data is described. Researchers use a variety of statistical analysis methods to analyze different single subject research designs. This dissertation presents a series of statistical models to model data from various conditions: the baseline phase, A-B design, A-B-A-B design, multiple baseline design, alternating treatments design, and changing criterion design. The change-point evaluation method can provide additional confirmation of causal effect of the treatment on target behavior. Software codes are provided as supplemental materials in the appendices. The applicability for the analyses is demonstrated using five examples from the SSD literature.

In single-case research design (SCRD), experimental control is demonstrated when the researcher’s application of an intervention, known as the independent variable, reliably produces a change in behavior, known as the dependent variable, and the change is not otherwise explained by confounding or extraneous variables. SCRD studies that fail to demonstrate experimental control may not be published because researchers may be unwilling to submit these papers for publication due to null findings and journals may be unwilling and unlikely to publish null outcomes (i.e., publication bias). The lack of submission and publication of null findings, leading to a disproportion of positive studies in the published research literature, is known as the “file drawer effect” (Rosenthal, 1979; Ferguson & Heene, 2012). Recently, researchers and policy organizations have identified evidence-based practices (EBPs) for children with autism spectrum disorder (ASD) based on systematic reviews of SCRD studies (Odom, Collet-Klingenberg, Rogers, & Hatton, 2010). However, if SCRD studies that do not demonstrate experimental control (i.e., null studies) are disproportionately unpublished due to the file drawer effect, this may result in a misrepresentation of positive findings, leading interventions to be deemed evidence-based that, actually, lack sufficient empirical support (Sham & Smith, 2014; Shadish, Zelinsky, Vevea, & Kratochwill, 2016). Social narratives, exercise, self-management, and response interruption/redirection are interventions for children with ASD that has been named EBPs according to the National Autism Standards (NAC; 2009) and National Professional Development Center on Autism Spectrum Disorder (NPDC; 2010); however, these interventions have not yet been evaluated for potential publication bias. The study employed and extended methods similar to Sham and Smith (2014), comparing the procedures and results of published articles and unpublished dissertations and theses for interventions identified as EBPs to evaluate the methodological rigor and evaluate the possibility of publication bias, file drawer effect, and lack of replication. Specifically, the results of published and unpublished studies were compared to determine if published studies showed greater treatment effect, which would indicate the file drawer effect. Also, SCRD quality indicators were employed to evaluate whether studies that were published tend to be of higher quality, as this would mitigate possible publication bias shown by larger effect sizes (ES) in published studies. The outcome resulted in three out of four EBPs (social narratives, antecedent exercise, and response interruption and redirection), yielding different ES when published studies were compared to unpublished studies; in contrast, self-management yielded a similar ES for published and unpublished studies. For social narratives and antecedent exercise, unpublished studies presented at lower estimated ES than published studies; whereas for response interruption and redirection, unpublished studies presented at a higher estimated ES compared to published studies. Generally, study quality presented at similar levels for published and unpublished studies for each EBP, with the exception of antecedent exercise. Differences were identified for antecedent exercise study quality based upon visual and statistical analyses. Lastly, there do not appear to be observed differences in treatment outcomes between published and unpublished studies when study quality was considered in the analysis. Implications of the results are discussed with respect to the file drawer effect and publication bias in EBPs, and the call to increase publications in peer-reviewed journals of negative findings and replication studies, which leads to identifying and establishing boundary criteria for EBPs. / Special Education

No description available.